Granted, a nontrivial difference is that the barrier to entry is lower; photo editing is something that requires active effort and learning.
I also think this is "art" in service of commerce. This is OpenAI advertising their goods using art/design/writing. That's no different than cereal companies using Elmer's glue instead of milk for their photoshoots. I don't have a high-bar for that kind of "art".
The good news is that the cutting edge of art will (for a while longer) still be a human domain. The more popular these models become, the more of their images we see in our lives, the more we will value things that look different.
This is the last place to get a reasonable take of how the average person feels about this stuff.
I acknowledge that I'm not particularly good at predicting the future, but I'm confident that AI is here to stay.
But you say yourself you "have to consciously remind [yourself]" it isn't real. The Uncanny Valley is not applicable when true subjective realness is imparted.
The uncanny valley is when it's just slightly imperfect which makes things feel "off".
When we've reached the point that the AI is indistinguishable from humans, we've exited the uncanny valley.
No point trying to reason with them.
These products don’t stick - just like sora they’re seemingly cool initially but then people go back to what were already doing ex-ante.